Privacy and personalisation: predicting Parkinson’s disease severity from real-world gait with federated learning
This exploratory study shows that our proposed approach enables FL to achieve similar accuracy to traditional Machine Learning without sharing any patient data but with costs to the local performance, leading towards a smart-home system that prioritises personalisation and patient privacy.
Authors
- Chloe Hinchliffe (ORCID: https://orcid.org/0000-0002-5002-1120)
- Paul Watson (ORCID: https://orcid.org/0000-0001-7629-4760)
- Hugo Hiden (ORCID: https://orcid.org/0000-0003-0843-5124)
- Silvia Del Din (ORCID: https://orcid.org/0000-0003-1154-4751)
- Alison Jane Yarnall (ORCID: https://orcid.org/0000-0002-3126-9163)
- Lisa Alcock
- Rachael A. Lawson
- Lynn Rochester
Institutions
- Newcastle upon Tyne Hospitals NHS Foundation Trust (GB)
- Clinical Research Institute (US)
- NIHR Newcastle Biomedical Research Centre (GB)
- University of Newcastle Australia (AU)
- Newcastle University (GB)
Publication Details
- Journal
- Frontiers in Aging Neuroscience
- Published
- 2026-03-09
- DOI
- https://doi.org/10.3389/fnagi.2026.1766599
- Primary Topic
- Parkinson's Disease Mechanisms and Treatments
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Newcastle University
- University of Wollongong
- Engineering and Physical Sciences Research Council